DESCRIPTION: Ionic channels are proteins of considerable interest, scientifically and medically. They control many types of transport across cell membranes and thus much of the life of cells and tissues, in health and disease. Every day hundreds of laboratories study the function of ionic channels one molecule at a time. The structures of a few channels are known, and more will be known in the next few years. The understanding of channels, in particular, the understanding of their function in terms of their structure, has lagged behind. Calculations of the properties of proteins are done now in many laboratories every day using techniques of molecular dynamics. But these calculations are severely limited in time span, unable to reach the biological time scale of channel activity which starts around 100 nsec. These calculations are based on direct integration of Newton's laws of motions, with interatomic forces described by empirical potential functions. Another formulation of Newton's laws, as variational principles, has been known for a long time, since Fermat or Lagrange. Many variational principles have been used for a variety of systems: e.g., Onsager & Machlup introduced one such principle for macroscopic, irreversible systems. Ionic channels and many proteins are such systems. They have shown that this principle can be applied to molecular systems such as proteins and that it allows a large increase in the time span of calculation. The investigators propose to use the OM action to calculate the properties of ionic channels. They will predict open channel currents, flickering, and noise. They will predict gating, particularly of variants of the native channel studied in detail in Andersen's laboratory. In this way, they hope to learn how gramicidin functions in atomic detail on a time scale approaching the biological.